Metabolic Modelling Reveals Evolution of Complex Phenotypes through Successions of Adaptive Steps

Comparing the metabolic networks of closely related bacterial strains can reveal their adaptive history. We analyzed 53 E. coli strains with published metabolic network models [1] and reconstructed their phylogeny. We considered each of the 52 internal nodes of the tree as an ancestral strain, and reconstructed the ancestral genomes and metabolic networks using maximum parsimony. We then identified the enzymes horizontally transferred between the ancestral strains and grouped them into segments according to their physical proximity in the 53 genome sequences.

We searched for the benefits brought by the transfer of those segments to the host metabolic networks, i.e., the new phenotypes acquired and the existing phenotypes enhanced, using flux balance analysis (implemented in the sybil R package [2]). We compared the benefits of those segments with a model that simulates the transfer of all possible genomic segments between different E. coli strains, as well as with another model that simulates transfer of randomized segments. Our analysis shows that, compared with randomized segments, real genomic segments are more likely to transfer reactions and phenotypes.

We further examined how a bacterium can acquire a complex phenotype, i.e., a phenotype whose genes are distributed across multiple locations on the genome, requiring transfer of more than a single segment. For each transferred or enhanced phenotype in a given strain, we searched for the contributing segments added since the strain’s most recent ancestor, and found that almost all phenotypic gains can be explained by the transfer of a single segment. However, if we trace back farther in time and identify the segments contributing to a given phenotype that were added along the phylogenetic lineage (i.e., after the most recent common ancestor of the 53 strains), then we found that a substantial fraction of phenotypic benefits (~40%) required multiple transfers. Thus, the most recent segment transfer required the presence of previous transfers to be beneficial. This is evidence of preadaptation (or “exaptation”): in order to adapt to an environment that requires complex metabolic changes, a bacterium may have to take  an evolutionary detour through “intermediate” environments to acquire segments one-by-one adaptively.

References

1. Jonathan M. Monk, Pep Charusanti, Ramy K. Aziz, et al. (2013). Genome-scale metabolic reconstructions of multiple Escherichia coli strains highlight strain-specific adaptations to nutritional environments. PNAS 110: 20338-20343
2. Gabriel Gelius-Dietrich, Abdelmoneim A. Desouki, Claus J. Fritzemeier, Martin J. Lercher, sybil - Efficient constraint-based modelling in R BMC Systems Biology, Vol. 7, No. 1. (13 November 2013), pp. 125-8, doi:10.1186/1752-0509-7-125